Correction of spatially variant phase error for synthetic aperture radar

1. A method for focusing an image, the method comprising: receiving an image comprising data collected from a synthetic aperture radar system; selecting a number of initial subpatches from the image, the number of initial subpatches not covering all of the image; estimating a phase error for each of the initial subpatches; creating a spatial model for spatial variation in phase error for the image based on the phase error for the initial subpatches, creating the model including determining a number of subpatches and the size of the number of subpatches, creating the spatial model including modeling a spatial version of the phase error for each of the initial subpatches using the phase error for the number of initial subpatches to form the spatial model; dividing the image into the number of subpatches based on the spatial model so as to cover the entire image with the number of subpatches; generating a number of coefficients for a polynomial for the phase error for each initial subpatch in the number of initial subpatches to form a number of sets of coefficients; fitting values for the number of sets of coefficients to form the spatial model; applying phase correction to each of the number of subpatches to form a number of focused subpatches, the applying phase correction including performing an optimization to identify an array of coefficients for a polynomial representing a phase correction in a manner that minimizes an entropy value generated by an entropy calculation for the image for each of the number of subpatches; and merging the number of focused subpatches to form a focused image.

2. The method of claim 1 , wherein the image is an initially processed image and further comprising: interpolating radar data to form interpolated data; performing a range inverse fast Fourier transform on the interpolated data to form transformed data; autofocusing the transformed data to form globally focused data; and performing an azimuth inverse fast Fourier transform on the globally focused data to form the initially processed image.

3. The method of claim 1 , wherein the dividing step comprises: identifying the number of subpatches using the spatial model; and dividing the image into the number of subpatches.

4. The method of claim 3 , wherein the step of identifying the number of subpatches using the spatial model comprises: obtaining total phase changes from all phase order errors in a range direction and an azimuth direction; and dividing the total phase changes by a desired phase bound.

5. The method of claim 1 , wherein the merging step comprises: performing one of combining overlapping areas in the number of focused subpatches using linear weighting or cutting and pasting to form the focused image.

6. The method of claim 1 , wherein the applying step further comprises: applying the array of coefficients for the each of the number of subpatches to the polynomial to obtain a desired phase correction for the each of the number of subpatches; and correcting a phase error in the each of the number of subpatches using the desired phase correction to focus the each of the number of subpatches.

7. The method of claim 1 , wherein the synthetic aperture radar system is located in an object selected from one of an aircraft, a spacecraft, and a satellite.

8. An apparatus comprising: an autofocus process capable of receiving an image comprising data collected from a synthetic aperture radar system; selecting a number of initial subpatches from the image, the number of initial subpatches not covering all of the image; estimating a phase error for each of the initial subpatches; creating a spatial model for spatial variation in phase error for an image based on the phase error for the initial subpatches, creating the model including determining a number of subpatches and the size of the number of subpatches including modeling a spatial version of the phase error for each of the number of initial subpatches to form the spatial model; generating a number of coefficients for a polynomial for the phase error for each initial subpatch in the number of initial subpatches to form a number of sets of coefficients; fitting values for the number of sets of coefficients to form the spatial model; dividing the image into the number of subpatches based on the spatial model so as to cover the entire image with the number of subpatches; applying phase correction for each of the number of subpatches to form a number of focused subpatches, the applying phase correction including performing an optimization to identify an array of coefficients for a polynomial representing a phase correction in a manner that minimizes an entropy value generated by an entropy calculation for the image for each of the number of subpatches; and merging the number of focused subpatches to form a focused image; and a computer, wherein the autofocus process executes on the computer.

9. The apparatus of claim 8 , wherein the image is an initially processed image, and wherein the autofocus process is further capable of interpolating radar data to form interpolated data; performing a range inverse fast Fourier transform on the interpolated data to form transformed data; autofocusing the transformed data to form globally focused data; and performing an azimuth inverse fast Fourier transform to form the initially processed image.

10. The apparatus of claim 8 , wherein in estimating the phase error for the number of initial subpatches, the autofocus process is capable of generating a number of coefficients for a polynomial for the phase error for each initial subpatch in the initial number of subpatches to form a number of sets of coefficients.

11. The apparatus of claim 10 , wherein in modeling the spatial version of the phase error for the number of initial subpatches to form the spatial model, the autofocus process is capable of fitting values for the number of sets of coefficients to form the spatial model.

12. The apparatus of claim 11 , wherein in dividing the image into the number of subpatches based on the spatial model, the autofocus process is capable of identifying the number of subpatches using the spatial model, and dividing the image into the number of subpatches.

13. The apparatus of claim 8 further comprising: a synthetic aperture radar system capable of generating radar data for the image.

14. The apparatus of claim 8 further comprising: a mobile platform, wherein the synthetic aperture radar system is located on the mobile platform.

15. The apparatus of claim 14 , wherein the mobile platform is selected from one of an aircraft, a spacecraft, and a satellite.

16. A non-transitory computer program product for focusing an image, the computer program product comprising: a computer recordable storage medium; program code, stored on the computer recordable storage medium, for receiving an image comprising data collected from a synthetic aperture radar system; program code, stored on the computer recordable storage medium, for selecting a number of initial subpatches from the image, the number of initial subpatches not covering all of the image; program code, stored on the computer recordable storage medium, for estimating a phase error for each of the initial subpatches; program code, stored on the computer recordable storage medium, for creating a spatial model for spatial variation in phase error for the image based on the phase error for the initial subpatches, creating the model including determining a number of subpatches and the size of the number of subpatches, creating the spatial model including modeling a spatial version of the phase error for each of the initial subpatches using the phase error for the number of initial subpatches to form the spatial model; program code, stored on the computer recordable storage medium, for dividing the image into the number of subpatches based on the spatial model so as to cover the entire image with the number of subpatches; code, stored on the computer recordable storage medium, for generating a number of coefficients for a polynomial for the phase error for each initial subpatch in the number of initial subpatches to form a number of sets of coefficients; code, stored on the computer recordable storage medium, for fitting values for the number of sets of coefficients to form the spatial model; program code, stored on the computer recordable storage medium, for applying phase correction to each of the number of subpatches to form a number of focused subpatches, the applying phase correction including performing an optimization to identify an array of coefficients for a polynomial representing a phase correction in a manner that minimizes an entropy value generated by an entropy calculation for the image for each of the number of subpatches; and program code, stored on the computer recordable storage medium, for merging the number of focused subpatches to form a focused image.

17. The computer program product of claim 16 , wherein the image is an initially processed image and further comprising: program code, stored on the computer recordable storage medium, for interpolating radar data to form interpolated data; program code, stored on the computer recordable storage medium, for performing a range inverse fast Fourier transform on the interpolated data to form transformed data; program code, stored on the computer recordable storage medium, for autofocusing the transformed data to form globally focused data; and program code, stored on the computer recordable storage medium, for performing an azimuth inverse fast Fourier transform on the globally focused data to form the initially processed image.